|
|
|
Uni Köln
> WiSo-Fakultät
> Seminar
für Wirtschafts- und Sozialstatistik > Institut
> LS
Mosler > Prof. Mosler > Datenportal
Datenportal des Lehrstuhls für Statistik und Ökonometrie
|
| |
Biomedical data
The data set (and description) can be downloaded here:
http://lib.stat.cmu.edu/datasets/biomed.data
Description:
February 23, 1982
The 1982 annual meetings of the American Statistical Association (ASA)
will be held August 16-19, 1982 in Cincinnati. At that meeting, the ASA
Committee on Statistical Graphics plans to sponsor an "Exposition of
Statistical Graphics Technology." The purpose of this activity is to
more fully inform the ASA membership about the capabilities and uses of
computer graphcis in statistical work. This letter is to invite you to
participate in the Exposition.
Attached is a set of biomedical data containing 209 observations (134
for "normals" and 75 for "carriers"). Each vendor of provider of
statistical graphics software participating in the Exposition is to
analyze these data using their software and to prepare tabular, graphical
and text output illustrating the use of graphics in these analyses and
summarizing their conclusions. The tabular and graphical materials must be
direct computer output from the statistical graphics software; the
textual descriptions and summaries need not be. The total display space
available to each participant at the meeting will be a standard poster-
board (approximately 4' x 2 1/2'). All entries will be displayed in one
location at the meetings, together with brief written commentary by
the committee summarizing the results of this activity.
Reference
Exposition of Statistical Graphics Technology,
L. H. Cox, M. M. Johnson, K. Kafadar,
ASA Proc Stat. Comp Section, 1982, pp 55-56.
Enclosures
THE DATA
The following data arose in a study to develop screening methods to
identify carriers of a rare genetic disorder. Four measurements m1,
m2, m3, m4 were made on blood samples. One of these, m1, has been used
before.
Because the disease is rare, there are only a few carriers of
the disease from whom data are available. The data come in two files,
one for normals and one for carriers of the disease. A description of
the files is provided. The data have been stripped of the names and
other identifiers. Otherwise the data are as received by the analyst.
PURPOSE OF THE ANALYSIS
The purpose of the analysis is to develop a screening procedure to
detect carriers and to describe its effectiveness. Experts in the
field have noted that young people tend to have higher measurements.
The laboratory which prepared the measurements is worried that there
may be a systematic drift over time in their measurement process.
These effects should be considered in the analysis. Can graphical
displays show the differences between the distributions of carriers
and normals?
FILE DESCRIPTION
Column Content
1 Observation number (sequence number per patient)
Note that there are several samples per patient
for some patients.
2-8 Blank
9-12 Hospital identification number for blood sample
13-18 Blank
19-20 Age of patient
21-26 Blank
27-32 Date that blood sample was taken (mmddyy)
Note that all day entries are 00.
33-39 Blank
40-43 ml (measurement 1) sss.s
44-50 Blank
51-54 m2 (measurement 2) xxx.x Eight missing data points.
55-61 Blank
62-65 m3 (measurement 3) xxx.x
66-72 Blank
73-75 m4 (measurement 4) xxx Seven missing data points.
Descriptive statistics:
Dataset= biomed : n= 194 , d= 4
Class1: n= 67
Covariance matrix:
[,1] [,2] [,3] [,4]
[1,] 51498.5920 -305.2888 2716.0509 7973.548
[2,] -305.2888 97.1740 -20.0220 -74.839
[3,] 2716.0509 -20.0220 296.2756 600.102
[4,] 7973.5479 -74.8390 600.1020 5247.087
Correlation matrix:
[,1] [,2] [,3] [,4]
[1,] 1.0000 -0.1365 0.6953 0.4851
[2,] -0.1365 1.0000 -0.1180 -0.1048
[3,] 0.6953 -0.1180 1.0000 0.4813
[4,] 0.4851 -0.1048 0.4813 1.0000
Median: 108.4149 91.6202 19.625 238.5618
Mean: 185.791 92.9313 23.9328 250.9403
MCD-estimated:
MDC-0.975-Mean: 82.2917 93.4854 17.3 234.6667
MDC-0.750-Mean: 82.2917 93.4854 17.3 234.6667
MDC-0.500-Mean: 82.2917 93.4854 17.3 234.6667
Class2: n= 127
Covariance matrix:
[,1] [,2] [,3] [,4]
[1,] 345.8414 -75.8980 8.3771 152.9563
[2,] -75.8980 149.7853 4.3577 92.4229
[3,] 8.3771 4.3577 19.3066 39.7680
[4,] 152.9563 92.4229 39.7680 1711.3575
Correlation matrix:
[,1] [,2] [,3] [,4]
[1,] 1.0000 -0.3335 0.1025 0.1988
[2,] -0.3335 1.0000 0.0810 0.1825
[3,] 0.1025 0.0810 1.0000 0.2188
[4,] 0.1988 0.1825 0.2188 1.0000
Median: 36.9543 82.9641 11.8769 160.7384
Mean: 39.6165 82.3071 12.1457 164.5748
MCD-estimated:
MDC-0.975-Mean: 33.4415 83.8075 12.0377 160.5
MDC-0.750-Mean: 33.4415 83.8075 12.0377 160.5
MDC-0.500-Mean: 33.4415 83.8075 12.0377 160.5
Measures:
Mah.Dist: 1.9215
Mah.Dist-MCD-0.975: 1.8959
Mah.Dist-MCD-0.750: 1.8959
Mah.Dist-MCD-0.500: 1.8959
Zuletzt geändert am 17.02.2013
|
|
|
|
